Part 1 Description
Today’s conversation isn’t quite a documentary, but it isn’t just your average interview either. So I’m labeling this a deep dive, and the topic is tokenized debt.
The Flippening Podcast is first and foremost for institutional investors, and I’m covering this topic because the dual issues of debt and lending are just too big to ignore and absolutely rife with investment opportunities. Many of which we talk about during this conversation.
This is the first of two episodes, in which we break down the topic of tokenized debt into five chapters. Today, you will hear the first few chapters of this conversation.
In chapter one, we discuss the history and origins of debt from the origin of the human species, up until around the year 2000. We also discuss the web 2.0 peer to peer lending solutions that preceded Dharma, such as prosper and lending club.
In the second chapter, we discuss the history and development of Crypto Asset Technology that has paved the way for programmable, tokenized debt. In chapter 3 we discuss we discuss how the Dharma protocol works and the 1st class operators within the system.
Part 2 Description
As a reminder, in chapter one we discussed the history of debt from the origin of the species to the year 2000. In chapter two, we explored the history and development of crypto asset technology that has paved the way for programmable, tokenized debt.
Today’s episode begins with the second half of chapter three of this conversation, where we discuss how the Dharma protocol works and the first class operators within the system.
In chapter four, we focus on the top use cases and investment opportunities for tokenized debt. Finally, in chapter five, we discuss Dharma as a corporate entity, how it works, and more.
Thanks To Our Sponsors
We’d like to take this opportunity to thank our two sponsors, REX BKC ETF and the Nomics API.
BKC is managed by Brian Kelly, a Cryptocurrency and Blockchain professional, CNBC contributor, Bitcoin author, and BK2Cents blog author.
Equity can add a different dimension to cryptocurrency investing. Just as gold miners need pans and wash-plants, cryptocurrency needs hardware, software, miners, and exchanges – Brain finds the companies trying to profit from the building the crypto infrastructure and puts them all in one fund, BKC. Please visit rexshares.com for more information on the ETF and to sign up for Brian Kelly’s blog – BK2Cents.
Now here’s the legal wording: Investors should carefully consider the Fund’s investment objectives, risk factors including Cryptocurrency & Blockchain technology risk, charges, and expenses before investing. This and other information can be found in the Fund’s prospectus at rexshares.com/bkc. Read it carefully before investing. BKC is not suitable for all investors. Not all companies BKC invests in may have significant exposure to blockchain or cryptocurrency although it’s the focus of the fund. The advisor’s judgments may be wrong and lead to loss of principal. Distributed by Foreside Fund Services, LLC.
The Nomics API offers squeaky clean and normalized primary source trade data offered through fast and modern endpoints. Instead of having to integrate with several exchange APIs of varying quality, you can get everything through one screaming fast fire hose. If you found that you or your developer have to spend too much time cleaning up and maintaining datasets, instead of identifying opportunities, or if you’re tired of interpolated data and want raw primary source trades delivered simply and consistently with top-notch support in SLAs, then check us out here.
- The origin story of the Dharma Protocol.
- The historic and macroeconomic forces that led to the early stages of debt.
- The birth of peer-to-peer lending on the internet.
- What kind of problems Dharma was built to solve.
- How peer-to-peer platforms like Prosper and Lending Club work behind the scenes.
- Why a borrower would choose to use a peer-to-peer lending service over more traditional options.
- How loans might be collateralized in the future with security tokens.
- Who the first class operators are that are working within the Dharma protocol.
Links Relevant To This Episode:
- Nadav Hollandar
- REX BKC ETF
- Brian Kelly
- Lending Club
- Y Combinator
- Debt: The First 5,000 Years by David Graeber
- Khan Academy
Terms Mentioned in the Episode:
Part 1 Transcript:
Clay Collins: Welcome to Flippening, the first and original podcast for full-time, professional and institutional cryptoinvestors. I’m your host, Clay Collins. Each week we discuss the cryptocurrency economy, new investment strategies for maximizing returns and stories from the front lines of financial disruption. Go to flippening.com to join our newsletter for cryptocurrency investors and find out just why this podcast is called Flippening.
Speaker 2: Clay Collins is the CEO of Nomics. All opinions expressed by Clay and podcast guests are solely their own opinion [00:00:30] and do not reflect the opinion of Nomics or any other company. This podcast is for informational and entertainment purposes only and should not be relied upon as the basis for investment decisions.
Clay Collins: Hey, this is Clay. Before we get started with what’s become one of my favorite episodes today, I wanted to invite anyone who’s listening to this who’s passionate about crypto-asset market data to reach out to me on our contact form. My company Nomics has a crypto-asset market API product and will soon be releasing some new features that we’d love to get your feedback on. If you’re [00:01:00] a developer, analyst or fund manager that relies heavily on trade and order book data from exchanges, please contact me by going to nomics.com, navigating to the contact us form and dropping me a line. You can also find me on our telegram group if that’s easier @nomicstelegram.com.
Finally, if you want to use what is hands down the best crypto-asset market data API available right now, go to nomicsapi.com. The API is 100% free and has no rate limits. Today’s conversation [00:01:30] isn’t quite a documentary, but it also isn’t your average interview. So I’m labeling this a deep dive and the topic is Programmable Peer-to-Peer Tokenized Debt. The Flippening podcast, this podcast, is first and foremost for institutional investors. I’m covering this topic because the dual issues of debt and lending are just too big to ignore and absolutely rife with investment opportunities, many of which we talk about during this conversation.
While there’s been a lot of talk from the blockchain space about tokenized stores of value like cash as [00:02:00] well as tokenized securities and commodities, the topic of tokenized debt has received such little coverage. This is unfortunate because worldwide debt markets are actually larger than stock markets and much, much larger than money supplies. I’ll use the example of the United States to illustrate. Here goes. The M2 money supply in the United States is 13.5 trillion. As a sidenote, M2 money supply includes not only liquid forms of money like cash, coins and checking accounts but also less liquid forms of value such as savings, deposits, [00:02:30] money mutual funds and other time deposits not easily transferable to physical money.
So we’ve got the US M2 money supply at $13.5 trillion. By contrast, the size of the United States stock market right now is about 30 trillion, so over double the size. But guess what’s even bigger than the 30 trillion dollar stock market? Well, the United States bond market is bigger with over 40 billion in debt outstanding. And that’s just debt represented by the bond market, [00:03:00] it doesn’t even include household debt and other forms of debt. So to summarize, we have US money supply at $13.5 trillion dollars, the stock market at $30 trillion and the bond market at $40 trillion dollars. So yeah, debt and debt markets are a huge, huge deal.
Furthermore, when we put our anthropologist and historian hats on, we find it very likely that humans issued credit and debt long before we adopted currency or coinage. Debt came before [00:03:30] money. My co-creator for this content is Nadav Hollander who is the founder and CEO of Dharma, a platform for building borderless lending products using programmable tokenized debt. Incidentally, Dharma does not have a token, so he’s not here to shill his coin or pump his market cap. We spend about half of our time talking about debt, its history, the nature of debt markets and peer-to-peer lending with web 2.0 technologies on platforms like Lending Club and Prosper.
The second half of this conversation is about the Dharma protocol itself, [00:04:00] how it works, use cases and opportunities for entrepreneurs and developers around the protocol and platform. We start this deep dive by learning a little bit about Nadav and the origin story of the Dharma protocol. The rest of this deep dive is broken into five chapters. In the first chapter, we discussed the history and origins of debt from the origin of the human species up until around the year 2000. In the second chapter, we discussed the history in development blockchain technologies that have paved [00:04:30] the way for programmable tokenized debt and the Dharma Protocol.
In Chapter three, we discuss how the Dharma protocol works in the first-class operators within the system. Chapter four focuses on the top use cases and investment opportunities for tokenized debt. This chapter is of particular interest for investors that are looking to generate a return on this space. In the fifth chapter, we discussed Dharma as a corporate entity, how it works, where it operates, et cetera. I should mention that this [00:05:00] podcast is made possible by REX BKC ETF. BKC is managed by Brian Kelly, a cryptocurrency and blockchain professional, CNBC contributor, Bitcoin author and BK2Cents blog author.
Equity can add a different dimension to cryptocurrency investing just as gold miners needed pans and wash plants. Cryptocurrency needs hardware, software miners and exchanges. Brian finds the companies trying to profit from the building of crypto infrastructure and puts them all in one fund, BKC. [00:05:30] Please visit rexshares.com for more information on the ETF and to sign up for Brian Kelly’s blog BK2Cents. Now, here’s the legal wording. Investors should carefully consider the funds, investment objectives, risk factors including cryptocurrency and blockchain technology risk, charges and expenses before investing. This and other information can be found in the funds prospectus at rexshares.com/bkc.
Read it carefully before investing. BKC is not suitable for all investors. Not all companies BKC invest in may have significant exposure to [00:06:00] blockchain or cryptocurrency although it is the focus of the fund. The advisors judgments may be wrong and lead to loss of principal distributed by Foreside Fund Services, LLC. Again, for more information that’s, rexshares.com. Nadav can you tell us about the origin story of the Dharma Protocol and [00:06:30] your founding of the project.
Nadav Hollander: Sure. First of all just is a little bit of context what Dharma is a protocol for borrowing and lending crypto assets on blockchains. Our basic core thesis is that there are a whole litany of interesting financial services that can be built on block chains that we believe are going to be fundamentally more globally accessible and transparent and fair. But those are probably a little bit of a ways out and all of [00:07:00] them sort of lie downstream of having a robust decentralized credit market in general. So what Dharma is building is essentially kind of the infrastructure for a decentralized credit market in which anybody in the world can engage in sort of like a peer-to-peer debt agreement that is entirely mediated by a smart contract.
So that’s Dharma but it didn’t necessarily start that way. I started thinking about essentially this concept back when I was an engineer at Coinbase in [00:07:30] 2016 and this was particularly a time in which Coinbase had started sort of really putting its weight behind Ethereum and the community around it. It was really exciting because a lot of people were working on interesting projects that were basically like, let’s create prediction markets but decentralized or let’s create DNS but decentralized.
So my kind of kernel of an idea that I was interested in was creating basically like lending club but decentralized. I ended up tinkering [00:08:00] with that for like six months, playing around with a lot of different models for it, trying to learn as much as I can about kind of where people had failed in going down similar paths in the past. Eventually left Coinbase and graduated school as well and ended up getting into YCombinator and starting to work on this idea.
Clay Collins: What’s the like intellectual underpinnings of your journey and how did you get here? Did you study economics in the past? Have you always been involved in debt markets? Did you play around with Lending Club [00:08:30] and Prosper and those types of platforms in the past? Maybe taking out a loan or given one. The average person just doesn’t get up someday and say, I think we can reinvent global debt markets. What was your personal journey?
Nadav Hollander: I’d be lying if I said that I had some sort of formal background or training in debt capital markets. I think it’s also like important to note that the way that Dharma started off was like much more narrowly scoped than what it is right now. In that like, I’d admit that like when I started [00:09:00] working on Dharma, it was very much like naively concept. It is like we want to have peer-to-peer micro loans that anybody in the world can take out. The basic kind of like archetypal image that I wanted to make come true was like anybody in the world that has an internet connection can go out and take a loan online, which right now isn’t necessarily the case. It’s not like if you’re in Karnataka, India or something like that. You can go and go on lendingclub.com and take out a loan.
So I was very much at the time thinking of it from the perspective of like narrowly like how can you do things like [00:09:30] Lending Club, i.e, like peer-to-peer consumer lending from the perspective of … In a decentralized manner like that is borderless and globally accessible. And in that process like that’s very much something that led me to have to learn a lot about how lending businesses work, on how peer-to-peer lending businesses in particular have kind of panned out over the past 10 years or so. I the process, I got exposed to a lot of what is sort of the like Rube Goldberg machinery that underlies [00:10:00] the financial system.
I can’t necessarily say that I was like seasoned in like the battlefields of Wall Street per se, but I think that in having hashed out a lot of different models for how this protocol could work and actually built out various kind of first iterations of it I think I learned a lot of … I think I earned some of my stripes in terms of learning how this kind of machinery works right now.
Clay Collins: Let’s kick off chapter one. In chapter one, we’re going to put on our historian hats and attempt to contextually [00:10:30] place programmable tokenized debt in what Dharma is doing. We do this by exploring debt from the origin of the human species up until around the year 2000. We also explore web 2.0 peer-to-peer lending platforms such as Lending Club and Prosper. Nadav, can you speak to the historic and macroeconomic forces that take us up to the year 2000 and the birth of peer-to-peer lending on the internet.
Nadav Hollander: I can try to give you a crude summary of probably one of the better treatises [00:11:00] on this. There’s this book called Debt: The First 5000 Years, is written by a guy named David Graeber. Basically, kind of the core thesis of this book is that people have this conception of how money and loans kind of like came about and it usually goes something like this. The theory is like in some British village somewhere, I was trying to like trade my goat for something and somebody else had a bunch of shoes and we had this issue because I don’t want the person’s shoes and he doesn’t want my goat but we both [00:11:30] want to get rid of like our items in some capacity and liquidate it.
They call this the so called double coincidence of wants problem. The theory goes that essentially this is how money came about. It was like it was the bridge asset for everything. David Graeber’s really interesting argument that he makes is that actually in a weird way, people were kind of essentially like keeping credit tabs on one another well before they were actually using like [00:12:00] hard currency to purchase things. So you can imagine it being like if you lived in a village and there was some sort of informal social system in which like people knew at a certain shop that you were owed a certain amount of money by one tradesman and you were owed and certain amount of money by another tradesman, and you yourself owed a certain amount of money to another person.
All that stuff was kind of tracked very informally in the argument that David Graeber makes is essentially that this predates money and in many ways actually created money, and that [00:12:30] like money was the instrument that was used to eventually settle these different debts and accounts. So it’s really interesting because you can actually … You can kind of go back, and then this is exactly what his book does, and you can kind of go back through history and see that like a lot of … There’s just like so many historical debt accounting systems that almost predate money in many cases and that additionally like so much of the way that we talk about things in the world, like so much of our language is actually embedded with terms [00:13:00] that have to do with like debts in general and especially like when it comes to like morality.
A lot of the ways that we talk about like morality is spoken in terms of debts. Like for instance you have a debt to your parents for having like birth to you. It’s kind of like honor thy father sort of things. Basically, long story short, I’m giving you a crude summary of this extremely long novel right now. But the point is that pretty much as long as humans have been around, we’ve been keeping tabs on who owes who what and framing kind of a lot of [00:13:30] like our social ranking in terms of debts as well, and framing our society around them.
What’s kind of fascinating is that what had happened in kind of the invention of the financial system call it, let’s just say the modern financial system started in 1800s, which is you can make an argument that’s more like the 1600s. But let’s just say for sake of argument right now is like 1809s or so. Modern financial system starts to get developed and then all of a sudden things like the telegraph come around and people have the means to communicate globally [00:14:00] a speed that they never had the ability to do prior.
You have kind of this creation or like this parallel emergence of a formalized financial system, where people are not necessarily just kind of like informally keeping tabs of debts one another but rather are using legal contracts to like encode this person owes this much money to that person. Or rather like this country owes this much money to this bank. Arguably, a lot of kind of the development that came in the 1800s and 1900s had been financed [00:14:30] by this kind of emerging financial system. The most notable example being like the railroads. This is what made the JP Morgan’s of the world so powerful is that they basically financed the railroad.
Essentially, like the modern financial system emerged in the 1800s, 1900s or so, and basically like became kind of the engine of growth behind like America and also like the world at large in many senses. You’d asked [00:15:00] me to kind of stop before 2000, so I’ll pause there, but I think 2000 is where things kind of take an interesting turn in particular with respect to the financial crisis.
Clay Collins: I guess I was thinking about around 2003 we start seeing a bit of peer-to-peer lending activity. It sounds like you have some thoughts about the role of the financial crisis and some events that, that catalyzed.
Nadav Hollander: Yeah, absolutely. A lot of people would argue that the ” [00:15:30] success” of the early peer-to-peer lending platforms, those were almost entirely catalyzed by the fact that the financial crisis came about. In that like all of a sudden it was like much, much harder for a lot of people to get loans from traditional banks. So there was a really compelling argument to be made that creating a platform where you didn’t have to tap into kind of the traditional financial system to get a loan was very, very compelling and could open up credit to a lot of people who at the time were hurting.
[00:16:00] Now, that kind of rosy picture of like, we’re going to dismantle the financial system and we’re going to kind of use the internet to now let people lend directly to one another. That rosy picture isn’t exactly how things played out. It turns out that when you look at companies like Lending Club and Prosper and literally thousands of these companies nowadays. When you look at these companies, they’re largely kind of just like web 2.0 storefronts for the existing financial system. [00:16:30] Lending Club being a great example and so far as like I may be miss-stating the percentages here, but something like 80% of the loan volume on Lending Club is institutional.
It’s literally just like hedge funds cutting these large like debt facilities or these debt funds essentially basically like underwriting these loans in a way that basically leaves like the bottom tier of all the loans for retail investors. Mind you like on the back end, these companies are largely operating at the same sort [00:17:00] of technical fidelity that most banks are as well. It’s basically all Excel behind the glitz and glamour. So what you kind of saw is that like rather than these platforms is really actually like changing the fundamental fabric of how lending worked in the modern financial system, rather they kind of just became sort of like a cosmetic facelift.
Clay Collins: What problems remain to be solved bringing us to the origin of Dharma with systems [00:17:30] like Prosper and Lending Club, who was sort of left out of that system and what were the biggest problems that they needed solved?
Nadav Hollander: To answer that question I kind of like to look at the way the internet has influenced the development of technologies in general, or rather how it’s influenced the development of different sectors of business. If you look at for instance like content, content is a great example of a industry that was just absolutely totally transformed [00:18:00] and in my opinion for the better by the advent of the internet. The way that kind of happened is pretty simple. It’s like previously, the distribution of content was like a relatively expensive thing. Like you needed to have television rights, you needed to have like a physical distribution mechanism for things like newspapers and that generally created barriers of entry that in general just made content a less competitive industry.
Clay Collins: Before we get too much into the problems that existed [00:18:30] with these initial kind of web 2.0 peer-to-peer lending solutions, let’s walk through what the experience is for someone using one of these platforms. My understanding of how it works is someone signs up for one of these systems, they describe the kind of loan that they’re looking for and they say something about what they’re going to use it for. They fill out a profile, maybe employment status, there’s a credit check involved and then someone is on the other side of that, [00:19:00] that vets them according to whatever criteria they have and will offer them the loan. What else is involved with those platforms?
Nadav Hollander: In theory, yes. But there’s actually like a whole rigmarole that happens behind the scenes enable this to happen in a legal way. The biggest factor that needs to be taken into account is that there are super heterogeneous banking and lending regulations across the United States, in different states in America. Like, if [00:19:30] you are a company based in Silicon Valley and you want to issue loans to potentially anyone in the 50 states, basically the best way to do that is to be a bank. To be a chartered bank because chartered banks at least in certain states have these things called blue sky laws. I don’t think I need to get too much into the specifics here but the point is that like from a regulatory perspective, these things become like infinitely easier when you’re actually a bank.
This funny thing kind of developed around peer-to- [00:20:00] peer lending where essentially what happens actually behind the scenes is that there is a bank that is issuing a loan at the end of the day on a peer-to-peer lending platform. You’re not actually taking a loan out directly from the pier on the other side. What ends up happening is like a bank, and usually it’s in like a place like Utah, is going to like issue you like a paper loan. That loan is going to get bought by Lending Club and then that loan is going to get fractionalized and sold [00:20:30] to the other side of the market.
So what you really have at the end of the day is something that is several steps removed from like a true peer-to-peer transaction and that … At least the argument that I would make is that it’s the “peer-to-peer” aspect of it becomes much more of a marketing plate than it is actually an instrument of adding efficiency to the system.
Clay Collins: It’s just kind of a egalitarian seeming wrapper around a very old and traditional and regulated system. [00:21:00] With this peer-to-peer system, who’s left out and what problems are yet to be solved with these traditional peer-to-peer systems?
Nadav Hollander: There’s a couple caveats here. I actually think that in America, in like Western countries, there’s not that many people that are left out. Actually if you look at credit conditions in the United States right now, they’re pretty good. Most people who have pretty good credit are able to take out a loan. So when I think about [00:21:30] the people who are “left out” it’s not necessarily people like you and me, it’s not really kind of like your typical American that fits into that category.
In my opinion, what I find to be much more compelling and interesting subset of people that are kind of locked out of the current financial system, I think about the “unbanked”. Estimates tend to vary on as to like how many people in the world lack access to banking or traditional financial services, but some estimates have pegged it as high as 2 billion, some [00:22:00] are kind of sub 1 billion but still in like the high hundreds of millions. Irrespective, we’re talking about a significant subset of the human population.
If you start to account for people who are, as we call them, under banked, who potentially have access to the financial system but their only options are like extremely [userís 00:22:21] or like tend to like take advantage of them in various capacities, then that number grows significantly. So it’s really these people that I [00:22:30] think are an interesting kind of customer segment that doesn’t necessarily have access to these wonderful new peer-to-peer lending sites. Because at the end of the day, if you’re somebody who is in like a village in Africa or somebody who is in Nepal or what-have-you, you may have a wonderful internet connection that allows you to go on YouTube and learn everything you want on Khan Academy.
But it’s not like you can like have that same sort of borderless access to [00:23:00] financial services because the underlying rails at the end of the day is paper loans and kind of like the traditional banking system, which has not evolved in the same way that the internet has. It’s not evolved in this sort of kind of like open borderless manner, it’s still kind of very much rooted in localized institutions and localized banking infrastructure.
Clay Collins: Can you help me understand why someone in the United States would opt to use a platform like [00:23:30] Prosper, Lending Club or one of these kajillion of things that exist out there versus going to their bank. Is it because the amount to be borrowed is just so small that it isn’t worth going to a bank. I see white institutions would do this. I know a lot of people who are, they like the hobbyist aspect of being involved with peer-to-peer lending and it’s a fun hobby for them. Why does a borrower choose something like Lending Club over other options that might exist?
Nadav Hollander: I probably [00:24:00] am going to be misrepresenting this a little bit because I don’t work at these companies, so I can’t tell you their exact customer profile. But through what I’ve understood, a lot of it is just like having like a web storefront for loans. That’s just something that’s like really nice. People like to access products through the web, it turns out, that has driven the creation of like thousands of companies and startups in the past couple years. A lot of it is just like, I want to get a loan and I want to go online and do it. I [00:24:30] don’t really care if there’s a peer on the other side who’s like paying me to do that. It’s just a lot more practical than walking into a bank branch. I think that’s a big component of it.
There is also a certain component of it that is essentially that certain people fall into a category where they don’t necessarily have the best credit scores. They are a relatively uncredited worthy person and this kind of creates a little bit of like an ad for selection effect where the sorts of people that are more likely to flock [00:25:00] to these sorts of like alternative lending solutions may not necessarily be the best borrowers. A great kind of example of how this can go awry is that in China right now, there’s an enormous amount of peer-to-peer lending companies out there, many of which are like really, really sketchy operations and they tend to attract like really, really bad and often fraudulent borrowers. It’s not the rosiest picture ever.
At the end of the day, at least in a country where there is a really strong financial system, [00:25:30] if you already have good options at your hands, your take them and you don’t really care about whether it’s a big bank giving you that loan or somebody else. I think there’s a little bit of a problem in that like there’s somewhat of an adverse selection effect often with many of these platforms.
Clay Collins: Getting back to Dharma and the big opportunity that presented itself to you, it seems like there’s a number of reasons for you to exist. One is that this open financial system [00:26:00] that’s being built right before our very eyes starting with the birth of Bitcoin really doesn’t have a native way to lend and borrow money. Then there’s the issue of the unbanked and the underbanked and generally the underserved people worldwide who don’t have access to debt markets. Is there anything else that I’m missing?
Nadav Hollander: Yeah. I think that there is something to be said about empowering entrepreneurs to start businesses [00:26:30] in the world of finance by significantly lowering the barrier to entry to building an application that in some way involves lending. Like if you wanted to … Let’s just say, we’ll go back to the kind of canonical guy in India example. If you’re if you are a person in India who has a great idea on how to originate and underwrite debt locally and you understand some subset of the credit market there in a way that nobody else does. [00:27:00] You know who the good borrowers are, you know how to tell if they’re good borrowers, et cetera.
If you wanted to go and start a lending business, the first thing you got to go do is raise the requisite, sort of debt view. You’ve got to go door-to-door in Mumbai ostensibly and kind of talk to a lot of traditional debt investors and try to convince them that you’re going to be good at that job. That’s obviously a very like onerous and bespoke process that kind of locks a lot of people out of that ability. The beauty of systems like Bitcoin in general and I would argue this is kind of what we’re [00:27:30] trying to bring to lending markets with Dharma, is that like all of a sudden, the barrier to entry is just like whether you can code. It’s pretty much just a matter of like, can you log onto the internet and deploy this application? If so, then you have like native access to this infrastructure.
Similarly, that’s kind of what we’re trying to do with Dharma. We want it to be the sort of thing where if you want to plug into this decentralized credit market and essentially start underwriting different loans and start putting loan offers onto the network, you can do that. All you need is essentially [00:28:00] just like a couple lines of code and run a server, basically. As opposed to having to go out and get like a banking license and go out and traipse around Mumbai or Manhattan or whatever to raise the necessary money. So I think there’s something exciting about the idea.
I think that lowering the barrier to entry to entrepreneurs who want to start financial services businesses fundamentally creates more competition for those businesses, which means that the prices of those businesses go down and [00:28:30] the quality generally tends to go up. I think that in a second-order effect means that the consumer ends up getting something better.
Clay Collins: So this isn’t just about creating a decentralized tokenized version of Lending Club, this is about creating a platform on top of which permissionless innovation can happen around loans and these kinds of debt markets. Is that fair?
Nadav Hollander: That’s a great characterization, yeah.
Clay Collins: Okay. So that concludes chapter one. [00:29:00] Let’s move along to chapter two, which is an exploration of the blockchain technologies upon which Dharma and programmable tokenized debt and peer-to-peer lending stand. In this chapter, we’re trying to understand the technology stack that paved the way for the Dharma protocol. Nadav, can you describe for us the crypto-asset projects that have set the foundation for Dharma.
Nadav Hollander: The first thing to talk about in any case which I’m sure many of your listeners are very familiar with is Bitcoin. What was really cool about Bitcoin [00:29:30] that kind of changed the whole game of what you could build in terms of financial applications is that, all of a sudden, you had payments as like a native programmable primitive that were totally borderless. So all of a sudden is like if you wanted to build an application that anybody in the world could use that in some way accepted payments, you could very easily whip that up by plugging into like a Bitcoin node. So that’s a really radical concept and obviously that kind of spurred the development of the [00:30:00] crypto spaces we know today.
The purpose of that little social contract is to go out and basically make it so that if the person doesn’t end up paying back [00:31:00] that you have some sort of social proof that you guys agreed on this thing, and then you can take that social proof and show it to a judge and they’re going to go and like whack the person on the head. More realistically, they’re going to go and seize on their property.
Clay Collins: Time out. I’m going to do some native advertising for the Nomics API. This episode of Flippening is sponsored by the Nomics API. The Nomics API offers squeaky, clean and normalized primary source trade data offered through fast and modern endpoints. Instead of having to integrate with a bunch of exchange APIs of varying quality, [00:31:30] you can get everything through one screaming fast fire hose. If you found that you or your developers have to spend too much time cleaning up and maintaining data sets instead of identifying opportunities or if you’re tired of interpolated data and want raw primary source trades delivered simply and consistently with top-notch support in SLAs, then check us out at nomics.com. Okay, back to the show.
Nadav Hollander: Essentially, what Ethereum enabled us to do that we couldn’t necessarily do with Bitcoin is to kind of programmatically [00:32:00] encode these agreements between two different parties, and also to like programmatically encode the sort of punitive actions that would take place if a borrower fails to repay. The best example of that in Dharma today is that most of the loans that are issued by Dharma protocol are collateralized. Meaning that essentially like if Alice and Bob want to engage in the debt transaction, Alice promises to put up a certain amount of a certain asset as collateral [00:32:30] in order to receive some other asset. They engage in that debt agreement and for the duration of their loan the kind of smart contract that represents that loan actually acts as the custodian of the collateral. It actually like holds on to the collateral on Alice’s behalf.
Essentially, the reason why something like in Ethereum becomes relevant here is because you need this sort of like kind of dynamic programmable functionality, where a contract can keep track of like complex terms like, has this person actually [00:33:00] paid back according to the schedule that they promised? Has this person actually put up the collateral that we asked them to, et cetera. Which were things that were highly non-trivial and not necessarily possible with Bitcoin, which had like a fairly simplistic and primitive scripting language.
Clay Collins: Thinking about some of these security tokenization platforms like Harbor and Polymath that exists that are built on top of Ethereum, you can imagine a scenario where someone issues several tokens [00:33:30] that represent ownership. In their home in that future scenario that doesn’t quite exist yet, they would be able to collateralize a loan with tokens that represent ownership of the home. There’s of course complications around what happens if there’s a default and can you actually possess someone’s house with these tokens.
But I think that gets to the idea around the future that could exist here. Is that the way that Dharma essentially works? Does every loan have to have collateral that underpins the value of the loan or [00:34:00] are there ways to do credit scoring and take a look at people’s history and track record and make assessments about whether or not someone’s worthy of that loan without collateralization?
Nadav Hollander: First of all, the way that we’ve built Dharma is that it’s like a super generic mechanism for issuing debt. So there’s pretty much like almost no hard and fast requirements for what needs to happen when a loan takes place. You could have a on chain collateralized loan or you could have an entirely uncollateralized loan [00:34:30] where the two parties just happen to know each other and like there’s a social reason to repay. First point is that we built this in a generic way on purpose so that like you could basically have this kind of unified credit market for an entirely diverse set of loan types.
Now, with respect to kind of like your question about are most loans collateralized or are all uncollateralized, are there ways in which you can have loans that are uncollateralized that have some sort of credit scoring attached to them, et cetera. Essentially [00:35:00] our approach has been to focus first on collateralized loans insofar as like there is a demand for those right now from essentially speculative borrowers, people who want to like either short things or leverage things or take on some sort of merge in it.
It is possible today to do unsecured lending in Dharma. We have a mechanism that we call like the underwriter mechanism where essentially you can have a third-party underwriter come in and effectively kind of co-sign on the borrower’s behalf [00:35:30] their creditworthiness in the lending agreement, there’s a guarantor of that lending agreement. But we also kind of underline that this is like a experimental feature set and not necessarily something that we think is going to be like the primary driver of liquidity in Dharma for a while. The reason being that it’s just like underwriting and reputation systems are very complex, especially in a highly adversarial environment like the crypto community. So while we think this is possible, this isn’t necessarily something that we are [00:36:00] like putting all of our energy into encouraging people to do right now.
Clay Collins: Let’s close out chapter two and transition to chapter three. In Chapter three, we explore how the Dharma protocol works and the first-class operators within the system. Nadav, when I think of traditional debt markets and loans, I think of a lender and a borrower first and foremost. I also think about identity systems, underwriters and credit checking systems. At a high level, who are the first-class operators working within the Dharma protocol?
Nadav Hollander: [00:36:30] I think like the way that you phrased it is actually a great phraseology. There are kind of these enshrined first-class operators. Basically, there are … Besides the borrower and the lender, which are pretty straightforward to understand, there are two important actors to be aware of in the system. The first is what we call a relayer. Now, for those of you who are in the audience who may be familiar with what’s called the 0x protocol, it’s kind of gained a lot of attention recently, this is a highly analogous concept. [00:37:00] Basically, what a relayer is, is some sort of centralized entity that hosts a public order book where people can post requests for loans and where others could kind of see those requests and fill those requests by becoming lenders.
The relayer’s incentive to do this is that the protocol has a built-in way of compensating the relayer with a fee when an order that was listed on their order book is filled by a creditor. [00:37:30] Essentially, the idea is that this is basically like a super light way of spinning up a peer-to-peer marketplace business using just like open source code and without having to have like a bank account. All you have to do is host a server with one of these order books and people can go and post orders onto there and get matched with other lenders and borrowers. That’s the role what a relayer plays. Essentially, these relayers are kind of like the primary drivers of liquidity.
So we’ve already seen two different relayers come to main [00:38:00] net in the two months that we’ve been live but there are many others that were aware of in the pipeline. If anybody in the audience is listening who thinks that this might be an appealing business for them to start exploring, then by all means reach out to us. Now, the second set of first-party actors that are important to talk about are the underwriters that I sort of touched on earlier. Basically, what underwriters do in Dharma is not necessarily any different than what underwriters do in the traditional financial system.
Basically, [00:38:30] an underwriter in Dharma takes a loan request from a borrower and makes an assessment of that borrower’s creditworthiness in the context of that loan. They basically make a prediction of like how likely do I think it is that this loan request will be eventually defaulted on? The idea is that like they have to kind of like cosign onto the loan now. They have to say that like, I underwriter X believe that loan Y has 2% chance of defaulting, and [00:39:00] crucially when the loan is actually filled, not only do they get paid some sort of like pre-specified fee, but that prediction also gets immutably engraved into the blockchain.
This creates this interesting sort of incentive system where if a underwriter is doing a particularly terrible job of underwriting loans, that’s plainly visible in the blockchain. Because you’ll be able to see that all the historical loans that they have attested to did not end up turning out well or did [00:39:30] not end up turning out how they had kind of predicted. Whereas the opposite is also true. Like if a underwriter is doing a particularly good job of underwriting different loans, than the empirical performance of their loan scoring should be evident as well.
Clay Collins: So both the underwriter and the borrower are potentially accumulating reputation within the system over time, is that correct?
Nadav Hollander: I wouldn’t go so far as to say that there’s like any sort of formalized reputation system here, but at the very least the [00:40:00] underwriter is to some degree. This isn’t like a end-to-end trustless mechanism, like there is … At the end of the day, you do need to place some degree of trust in the underwriter, but we would argue that this is kind of better than the status quo insofar as like if you are investing in loans that a certain online lending website has originate underwritten, you’re kind of taking them at their word that the default rates that they’ve told you about are truly accurate. And that like [00:40:30] their books aren’t fraudulent and that they have like some degree of strong historical performance.
Whereas in Dharma, the underwriters which are functionally kind of performing the same role as one of these online lending websites. Like they’re basically originating the loan and credit scoring the borrower and credit scoring his likely to default. There’s actually an empirical signal that you can see that shows you how the cash flows actually panned out in comparison to loans they’ve underwritten it in the past.
Clay Collins: In many traditional [00:41:00] lending scenarios, the lender is also the underwriter. Is it fair to say that in this sort of system that you propose, the lending function and the underwriting function and be but don’t have to be unbundled?
Nadav Hollander: Yeah. That’s entirely accurate. I think that’s like part of our kind of core thesis, is basically taking what used to be kind of like inexorably wedged together in the lending process and kind of like stripping into its component pieces. You don’t have to do all these things separately [00:41:30] but if you want to be an underwriter, an underwriter only or if you want to be a lender and a lender only, you can do that.
Clay Collins: I think a lot of our listeners probably understand exchanges and cryptocurrency exchanges at least at a surface level better than they understand debt systems. It sounds like there’s definitely analogy here to an exchange. You’ve got the maker and there’s the taker, which is like the lender and the borrower. You’ve got an order book that’s provided by the relayer [00:42:00] and then you’ve got some kind of transaction fee, which is interest on the loan. I hope this isn’t offensive to say that this in a lot of ways sounds like zero acts for debt scenarios.
Nadav Hollander: That’s not offensive. I often characterize Dharma like that like a 0x for loans. Kind of is that. At the end of the day, because it’s basically like what a loan is basically like the borrower selling a debt asset to someone. Like, I’m going to sell [00:42:30] you the rights to these future cash flows at the price that is the current principle or below. So it is basically a maker and taker arrangement. Just the crucial things that need to be layered on top of it is like, how’s the taker going to have any idea of like how to price this thing in general. And so like there needs to be some sort of like underwriting mechanism or some sort of guarantees about the repayments to happen.
Two, just like one thing that is not possible in 0x which is inherently why [00:43:00] we need like an entire new set of infrastructure to do this at Dharma, is that you can’t post orders for assets that don’t exist yet. Then in the case of Dharma, that’s basically what it is. You’re posting an order saying, if you agree to this, I’m going to create a debt token and sell it to you for this price of the principle. That’s essentially the missing layer there.
Clay Collins: [00:43:30] We’re going to cut this conversation off right here. This concludes part one of this deep dive on tokenized programmable debt. Part two is just as long as today’s episode and just as good, perhaps even better. In part two, we complete chapter three then start and conclude chapters four and five which are about the top use cases in investment opportunities for tokenized debt and Dharma as a corporate identity respectively. It’s a fascinating conversation and a respectable conclusion to this deep dive on tokenized debt. As [00:44:00] a reminder, if you want to discuss any of this, you can join us on our telegram group @nomicstelegram.com. See you next week.
That’s it for this week. To sign up for our free crypto investing newsletter, listen to other episodes or get the show notes from this episode, please visit flippening.com. I also invite you to check out the startup that funds this podcast Nomics, spelled N-O-M-I-C-S, at nomics.com. Finally, if you got value from the show, the biggest thing you can do to help us out is to leave a five [00:44:30] star review with some comments and feedback on iTunes, Stitcher or wherever you listen to podcasts. Thanks for listening and see you next week.
Part 2 Transcript:
Clay Collins: 00:03 Welcome to Flippening, the first and original podcast for full time professional and institutional crypto investors. I’m your host, Clay Collins. Each week we discuss the cryptocurrency economy, new investment strategies for maximizing returns and stories from the front lines of financial disruption. Go to flippening.com to join our newsletter for cryptocurrency investors and find out just why this podcast is called Flippening.
Speaker 1: 00:25 Clay Collins is the CEO of Nomics. All opinions expressed by Clay and podcast guests are solely their own opinions and do not reflect the opinion of Nomics or any other company. This podcast is for informational and entertainment purposes only and should not be relied upon as a basis for investment decisions.
Clay Collins: 00:41 Hey, this is Clay and it’s a beautiful morning here in Minneapolis, Minnesota. It’s about 72 degrees outside. We’re about to kick off part two of this deep dive on programmable tokenized debt and debt markets with Nadav Hollander from the Dharma Protocol. Before we get started however, I want to thank you and everyone else who’s made what we’re doing here at Nomics possible. About a year ago, we started all of this with nothing. Now we have paying customers, we’re approaching being cash flow positive. We have sponsors and this dream that my co-founder and I had about this time last year is becoming a reality. Thanks to everyone who subscribed to our API or to this podcast on iTunes who sponsored a show, who shared our content on Twitter, who’s built apps with our data and linked back to us. Thanks to everyone who’s left a review for the show on iTunes or subscribed on iTunes or anyone else who’s referred this podcast or nomics.com to a friend or who’s mentioned us on Reddit or Telegram or wherever you share this stuff. Thank you for helping us make this dream we had a year ago a reality. Like you, we’re excited about the decentralization and tokenization of the world and we’re really thrilled that we get to do this full time. All right. I’m done being sappy. Let’s get to the show.
Clay Collins: 01:55 Welcome to part two of our deep dive on tokenized debt with Nadav Hollander from the Dharma Protocol. Part one of this series holds the record for the greatest number of downloads within 48 hours of publication. Like I said previously, this conversation isn’t quite a full fledge documentary, but it also isn’t your average interview either. So I’m calling this a deep dive. In part one of this conversation we discussed why debt and particularly tokenized debt is such a big deal. As a reminder, in the United States the money supply is valued at 13.5 trillion dollars. The stock market on the other hand is valued at 30 trillion dollars and the bond market is at 40 trillion dollars, and the bond market doesn’t even include household or other forms of debt. So it’s safe to say that in the United States, debt markets are much larger than the stock market and at least three times larger than the money supply.
Clay Collins: 02:43 So this conversation, the one that we’re having right now, is a large and important discussion and it deserves coverage. For more background and context around this discussion, please stop now and listen to part one of this series if you haven’t already. As an overview, this deep dive is broken into five chapters. In the first chapter, we discussed the history and origins of debt up until around the year 2000. We also discussed the Web 2.0 peer to peer lending solutions that preceded Dharma, such as Prosper and Lending Club. In the second chapter, we discussed the history and development of crypto asset technology that has paved the way for programmable tokenized debt. In chapter three, which we covered in part in the previous episode and finish up here, we discuss how the Dharma Protocol works and the first class operators within the system. Chapter four, which is the main focus of this episode, looks at the top use cases in investment opportunities for tokenized debt. In chapter five, we wrap everything up by discussing Dharma as a corporate entity, where they’re located, how they make money, stuff like that. In the previous installment of this deep dive that we covered, chapters one, two, and part of chapter three. In this second installment, we wrap up chapter three in complete chapters four and five. Let’s get to it. Nadav, thanks for sticking around for part two of this conversation and going the distance with me. Let’s get right into this and assume folks have listened to part one. Okay. So I think a lot of our listeners understand crypto exchanges better than they understand debt markets, at least most people in this space do. But it definitely seems like crypto exchanges can be used analogously to help us understand debt markets. For example, on crypto exchanges, you have a seller and a buyer who post a bid and an ask respectively. This is analogous to a lender and a borrower in debt markets. Also in debt markets, you have an order book of borrowers asking for loans and these orders are filled by lenders. With Dharma relayers, there can also be transaction fees which obviously go to the relayer and then there’s interest on the loan, which is also a factor. I hope this isn’t offensive and I mean this as a compliment truly, but it sounds a lot like the Dharma Protocol is basically Zero X for debt markets.
Nadav Hollander: 05:03 That’s not offensive. I often characterize Dharma like that, like as Zero X for loans. Like it kind of is that, right? Like at the end of the day, because it’s basically like what a loan is, it’s basically like the borrower selling a debt asset to someone. Right? It’s basically like I’m gonna sell you the rights to these future cash flows at the price that is the current principal below. So it is basically a maker and taper arrangement. Just the crucial things that need to be layered on op of it is like, “Okay. Well how’s the [inaudible 00:05:34] gonna have any idea of what like how to price this thing in general?” And so like there needs to be some sort of underwriting mechanism or some sort of guarantees about the repayments to happen. And two, just like one thing that is not possible in Zero X, which is inherently why we need like an entire new set of infrastructure to do this with Dharma, is that you can’t like post orders for assets that don’t exist yet. In the case of Dharma, like that’s basically what it is. Right? You’re posting in order saying like, “If you agree to this, I’m gonna create a debt token and sell it to you for this price of the principal.” That’s essentially the missing layer there.
Clay Collins: 06:10 It sounds like maybe second class operators or … I don’t know if operators are right, but like sort of things that exist, entities within the system are potentially identity systems and Oracle’s. I guess in a lot of cases those things are bundled that if you have a trustworthy idea of the real world identity of someone, now you can start doing traditional credit scoring. It’s hard to verify that someone owns a home or has a job without an Oracle system. But how do you think about identities in Oracle’s?
Nadav Hollander: 06:43 That’s kinda like the holy grail here, right? Like it’s like I mentioned earlier that there’s not really good reputation systems right now in the decentralized space. And something that I view as being a crucial element to enable that, is just like having a good identity standard. And there’s been lots of people who have been working on it for a while now and at least my opinion is that we haven’t really seen like any sort of meaningful traction yet in that regard. But I think that it’ll come eventually and when it does, I mean that’s gonna be hugely empowering, right? Because now all of a sudden you can actually have a … Now that like a public key is at least associated with someone with some sort of like identity that is not reproducible or not like … The technical term for this is a civil attack. Right? Like you can’t take it and just create like 7,000 copies of it. Once you have that, then you can actually build an on chain record of your credit worthiness. And so essentially you can imagine … I’ll give you an example really quickly from the real world.
Nadav Hollander: 07:40 So there’s a company called Branch that was started by a guy named Matt Flannery who originally started Kiva, which is the first micro lending platform that was on the internet. And essentially Branch, the way that they do … What they do is they give out basically like micro loans by via mobile phones in the developing world. And as far as I’ve understood a lot of the ways in which they give out loans to borrowers who basically have no credit history, is that they’ll basically start with like a really small amount. Right? Like so you can imagine I’m gonna lend you like $20.00 and you’re gonna owe me $21.00 in a week. And then once you’ve returned that small amount of money, probably 30% maybe, like 40% … I don’t know, some large percent of the people who got that initial amount of money just totally ran away with it and didn’t pay it back, but they’re not gonna get any other loans. Right? The X number of people who actually did pay it back, paid back the loan, and then now can take out a large loan or they can take out a $40.00 loan and so on and so forth until they kind of like build their credit worthiness overtime.
Nadav Hollander: 08:35 And surprisingly this model actually works, right? Like there’s definitely a certain portion of people who exit scammed the system and basically run off with some people’s money but by and large, the aggregate math of it works out that if the interest rates are set at a sufficiently high rate, that this can end up being profitable for the lender. So essentially what you can imagine is that if we had a good identity system on block chains, which right now we don’t necessarily, but imagine we did, then you could build that same sort of credit worthiness on chain in a decentralized manner. So you wouldn’t even need like an “underwriter” to sit there and judge the borrower’s credit worthiness, ’cause the borrower can just go up to a lender and basically say like, “Look, this is my public key associated with my immutable identity. And if you look at all the loans that have been taken out by public keys associated with my identity, I have this stellar credit worthiness that has taken me years to build up.” And so I really think that it’s crucial for us to like find a workable identity solution. I don’t wanna sound cynical about this. I have many friends who are in the industry who are working on this in various capacities and I believe in them and I believe in their projects. But we haven’t yet seen anybody like crack it successfully and really achieve meaningful traction.
Clay Collins: 09:49 How do you think about a potential skin in the game issue around the underwriters? If there isn’t an identity attached to the underwriter, they could do a stellar job approving these loans or denying and then all of a sudden approve just an incredibly large loan and then disappear. Can underwriters stake tokens or somehow put real world assets on the line?
Nadav Hollander: 10:12 So you’re really getting at the core of the reason why we consider unsecured lending to be a very experimental feature set in Dharma right now. And that’s basically that like you really need to put a lot of trust in the underwriter and in the system right now because not only can they do things like what you just kind of described, but they can mess with the whole system and a lot of more nefarious ways. Like a really bad example of this would be an underwriter going and basically like generating a bunch of different public keys and lending themselves a bunch of money and underwriting those loans. And basically like building up a false sense that they were a really good underwriter because they’ve underwritten all of these loans in the past. But when you look at the loans themselves, it’s actually just like all the underwriter’s public keys, and it’s just him lending money to himself. And then all of a sudden they go out and take out a giant loan and say that perfect credit worthiness, and then they run away.
Nadav Hollander: 11:04 And so … And this is a core reason why we say like this is very much like a delegation of trust. Like you’re pretty much just like … If you’re gonna be like investing in a loan that is underwritten by some party, like you better know who that party is and they better be like a public known figure. Like it better be … The example that I give is that like you probably wouldn’t want to invest in a loan that was just like some random public key that may have a good reputation, but it’s a different decision calculus if that like is … Coinbase is public key or something like that. Right?
Nadav Hollander: 11:32 So that’s kind of one example that I [inaudible 00:11:36] … Just to take a step back. We have been thinking a lot about like how can we make underwriters or rather the underwriting scheme at large more secure against these issues of them basically not having like a formalized reputation or identity. And we actually just recently wrote a blog post about this that I encourage people to read if they wanna go deeper into this, but basically there’s a couple of approaches that we’re thinking of. One is actually kind of adopting a structure that is very similar to the way that the Internet’s certificate infrastructure works and essentially like delegating trust between different underwriters. And to kind of dive a little deeper into that, I’ll give you kind of a specific example.
Nadav Hollander: 12:17 Imagine like we zoom like 20 years in the future or something like that and there are like two very well known underwriters in America that like have an excellent social reputation. Like they’re kind of like the equivalence of like an Experian and a TransUnion or something like that. Right? They’re just like these like household names, are super regulated, that if they were to try to pull off some sort of massive exit scam or something like that, it would just be like a massive public issue. Right? And like these companies would have to face some sort of legal repercussion and that’s great, but that’s not really like the world that we’re trying to aspire to. Right? Like we want to create a world in which there’s infinity different underwriters and that the scale’s in a much more decentralized way.
Nadav Hollander: 12:58 So what you can imagine instead is that you adopt a system that’s kind of like similar to the internet infrastructure of certificates, where like the way that certificates work on the internet right now is that there are these like root certificate authorities that are often for profit companies that are centralized entities, but like are a very limited set of entities. Right? It’s like some set of root level certificate authorities that all the browsers decide like we’re just gonna trust these root authorities. And then the way that you make that more scalable and make it so that you can have something as widely deployed as the internet and not everybody has to go through these root authorities is that these root authorities now can like vouch for other authorities. So they go out and they can vouch for a bunch of other certificate authorities and those certificate authorities can now go and vouch for other certificate authorities and et cetera, et cetera, and there’s this kind of chain of trust that gets formed.
Nadav Hollander: 13:54 And so the idea there is that when you visit a website and you see the little green lock on your site, essentially what’s happening is that your browser is checking the certificate that came with that website. That certificate was probably signed by like some random certificate authority. It doesn’t have to be like one of these root level ones. But if you were to go up the chain of trust, you would see that that certificate authority at the end of the day was like … it had up at … Like at the root of this chain of trust was like some known route authority. One thing that we’ve been exploring is kind of taking a similar model with underwriters, is essentially having underwriters kind of co-vouch for each other. So you can imagine there’s like a root level underwriter that’s like our TransUnion sort of equivalent and every sort of underwriting prediction that happens at the leaves, like some random underwriter in Arkansas or something like that, has sort of a chain of trust that delegates up to that root underwriter that kind of co-vouches for their underwritten prediction. And so in that sense, you kind of … you take this notion of starting with a small trusted set of underwriters and then take that and like chart it out. So now you can have many different underwriters making all kind of derived … their trust …
Nadav Hollander: 15:00 [inaudible 00:15:00] out so how you can have many different underwriters and they can all derive their trustworthiness from that small set. That’s one idea that we’ve been actively exploring.
Clay Collins: 15:08 I can see how that might work pretty well. Say there’s a country within that country, there’s regions within those regions. Maybe there’s little villages, and there might be a village leader who has a really great reputation and they can strike out on their own and try and be the country’s authority, or they can say, “I’m very much trusted within this village. Everyone knows me. I walk by so-and-so’s house every day. I have a sense of how many goats they have and what’s going on. I’m going to personally underwrite these,” and it rolls up. Yeah, I definitely can see that playing out.
Nadav Hollander: 15:42 These are early concepts like, for the more technical listeners out there, you can probably think of all sorts of technical flaws of this and all sorts of issues. This is very much something that’s a quarter-baked idea that we’re working on. But it’s one of the concepts that we’ve been playing with that we’re pretty excited about. The honest truth is that we’re not going to start throwing our resources into these things until we see how underwriting plays out empirically with the system that we’ve built so far, taking more of an experimental, empirical approach rather than trying to bequeath the perfect system from day one at the perspective of what we have right now.
Clay Collins: 16:14 Sure, yeah. We could hash out the … I guess this isn’t traditional crypto economics. This is crypto debt economics or something, but there’s a lot of game theory to play out in this field.
Nadav Hollander: 16:26 It’s super interesting that you say it that way, actually, because I was just having a conversation with Jill Carlson, actually, if you know who she is. What we were discussing is how there’s different standards of security that … I’ll take a step back. Jill comes from a background of trading sovereign debt at Goldman Sachs and then eventually later got into the crypto space. Something we were discussing is that it’s like the standard that crypto economists hold to security is usually that of cryptographers, right? Where a cryptographer’s standard is going to be like, “This hash function can’t be broken, even if we had a computer the size of the moon.” That’s great. That’s what you want. That’s what makes computer security as stable as it is.
Nadav Hollander: 17:08 But the irony is that inherently when it comes to debt capital markets, there’s no such thing as trustless because inherently the point is that you’re buying risk. You’re just pricing and buying risk at the end of the day. So you kind of have to take very different standards. The problem you’re solving is less so trying to eliminate the risk, per se, so much as just remove information asymmetries as much as possible. So it’s a very different problem set with very different constraints around it than, say, a reputation system in the context of a crypto economic system.
Clay Collins: 17:41 It seems somewhat analogous to hashing through the game theory related to democratic systems and voting. You assume some people aren’t going to vote. You can assume apathy within debt systems, you can assume people are going to default, you can assume that just bad things are going to happen and that’s built into the models. Yeah, it’s really interesting.
Nadav Hollander: 18:03 Not binary in success.
Clay Collins: 18:05 Right. I guess drilling a little bit deeper into this third chapter about what problems are being solved for whom, I’m really glad that we broke out some of these first class operators within the system and some of the related issues around identity and trust. At sort of a high level, what large problems do you think are being solved for whom? I can tell you, as a holder of crypto assets, I would love to get a return on my BTC denominated in BTC. I’d love to get a return on my denominated in ETH. That strikes me as a lender, that’s a problem. As a borrower, that use case is pretty well understood, the reasons why people borrow money. And then there’s developers and entrepreneurs who want the ability to solve problems in kind of a permissionless way, and so I guess that’s sort of the third category.
Clay Collins: 18:58 Something that strikes me here is that there’s a real tax, at least in the United States, to being centralized. That tax is that once you are centralized, the government can peer into your systems. If you’re a relayer, they’re not going to knock down your door because you’re not holding capital. But the second you’re custodying funds, a very different set of standards apply. I’m not thinking about this so much as people trying to evade regulation, but it does seem like once this plays out in a truly peer-to-peer decentralized fashion, that that frees up the innovators to truly innovate and not spend the bulk of their time filling out paperwork and talking to various government bodies.
Nadav Hollander: 19:42 Yeah. This is part of the thorny issue with any sort of protocol that is, as I would call it, semi-decentralized. It’s not fully end-to-end decentralized in that there are no centralized entities involved, but it’s an open market for centralized entities where, if one gets shut down, it’s really easy to crop up another one. With respect to the question of regulatory issues, if I’ve interpreted it correctly, the first thing I’d clarify is that relayers don’t custody any funds, ever, at any point. They are simply a bulletin board where these loan requests are actually listed.
Clay Collins: 20:21 Are they on chain or off chain books?
Nadav Hollander: 20:24 They could be both, right? I think most are going to be off chain because that’s way more efficient, but you could theoretically have an on chain order that was entirely decentralized. It would just be a lot more expensive. So they’re not necessarily custodying any funds, per se. With that being said, these are legal murky waters that nobody has answers for yet. You can ask any relayer, either in the 0X space or in the Dharma space, and they’re all trying their very hardest to be compliant but there’s really just not enough clear guidance yet around what legal classification these entities have. But I think it all just boils down, that goes back to the question that you’re asking, which is, okay, well, sure, the barrier may be lower. But now there’s all this other crap that you have to deal with in terms of the legal issues around it.
Nadav Hollander: 21:12 The response that I would have to that is twofold. First of all, I would argue that it creates dimensions for differentiation on both sides of the spectrum. So whereas one would think that purely economic theory, being a relayer should be a super commoditized bulk good, right? It should basically be like anybody can host one of these servers, and now it’s the same underlying technology, it’s the same service for the most part so the price should theoretically go down to zero.
Nadav Hollander: 21:41 But it’s actually, the regulatory friction points are kind of exactly where, in my opinion, the value capture is to be made for different relayers in that you can differentiate yourself on either side of that spectrum. The one very nefarious side of that is you could choose to be the Binance of relayers and just completely brazenly not care about regulations and list everything under the sun and enable huge amounts of liquidity in the ecosystem at a much higher regulatory risk.
Nadav Hollander: 22:13 On the other side of the spectrum, though, I would argue that you actually have the ability, for instance, for people to take an entirely regulated approach and cater instead to institutional buyers in the emerging market of the so-called security tokens, right? That in and of itself is something that is definitely going to be more expensive and is definitely going to take time and lawyers and all sorts of things. Those expenses that are spent upfront trying to construct a legally compliant relayer are also a differentiator and also a moat.
Nadav Hollander: 22:40 There’s an argument to be made that that is the side of the industry that’s going to have all the real liquidity on it. In the idealized version of what we’re trying to build, this could be entirely decentralized and you wouldn’t actually need to have centralizes relayer entities. But I think the more realistic stopgap solution that I think still achieves a lot of the same aims is having decentralized entities serving as relayers. I would argue that, even though there are legal complexities around it, that you can actually turn those to your advantage as a differentiator in a sense.
Clay Collins: 23:14 We see that playing out in all kinds of different forms with exchanges, right? You can go the Binance route and lean on jurisdictional arbitrage, or you can go the coin-based route and decide that you want to be regulated seven ways to Sunday and that you’re going to do everything in a fully compliant way. There’s definitely pros and cons to both paths, but that’s absolutely a place for potential differentiation. Is there value from a regulatory perspective in unbundling the underwriting function from the lending function? In other words, are you held to a higher standard if you are both the underwriter and the lender?
Nadav Hollander: 23:53 That is a question that I don’t feel comfortable answering. It’s just complex from a legal standpoint, and I don’t want to give bad legal advice.
Clay Collins: 24:03 So if you’re considering starting a relayer and you want to explore also being the underwriter, consult your attorney.
Nadav Hollander: 24:10 I would actually emphasize that much more forcefully. If you’re looking to build any sort of business within the Dharma ecosystem or, mind you, as a blockchain entrepreneur at large, make sure you have a lawyer and make sure that you’re consulting a good one because this is the murkiest sort of gray area of the legal space right now that’s at the intersection of fintech, which is already kind of a regulatory mess in and of itself, and cryptocurrency, which is like the wild, wild West badlands. So I’d emphasize it’s important to make sure that you’re doing things in a compliant manner.
Clay Collins: 24:44 Okay. That’s all for Chapter 3. Let’s move to Chapter 4. In Chapter 4, we’re going to focus on the top use cases and investment opportunities for tokenized debt. Chapter 4 is of particular importance to me because this podcast is first and foremost for institutional investors, and I’m covering this topic because the dual issues of debt and lending are just too big to ignore and absolutely rife with investment opportunities, many of which we’re going to talk abut right now.
Clay Collins: 25:10 Nadav, let’s talk about the top use cases for the Dharma Protocol. It strikes me that you’re somewhat of a philanthropist, not to say that you’re not a capitalist as well and that there isn’t economic gain to be had with your work on the Dharma Protocol, but you could’ve done a token sale, and you haven’t. I’m sure there are many businesses that you could create on top of your platform, but you haven’t done that yet either.
Clay Collins: 25:32 My question to you is this. When you put your capitalist hat on and you see all the opportunity exists within this emerging Dharma ecosystem, where do you see the most opportunity? And, if you were an entrepreneur or an investor listening to this podcast, what types of projects would you fund or build? In other words, when you see the range of opportunity that exists and consider an incredible number of potential use cases, which do you find to be the most compelling?
Nadav Hollander: 25:57 It’s funny that you ask this, because we’re literally about to release a blog post that we’ve called the Request for Blockchain Finance Startups, which is a play on, if you’ve ever followed YCombinator, they have their Request for Startups, essentially. We have all these ideas for things that we want people to build on Dharma that we think are going to be fantastic businesses. Essentially, we really want to seed the community with these ideas and see people build them out and build successful businesses.
Nadav Hollander: 26:23 I can give you a couple of examples of this. Unsurprisingly, they fall into the buckets of different types of relayers and different types of underwriters, because that’s like the first-class citizens of Dharma Protocol that, basically, we’re optimized for right now. Of the former, I’d say that there’s a really interesting model for creating basically a prime brokerage for security tokens. We discussed this earlier, but right now there’s a lot of people who are buzzing about this notion of security tokens and how huge they’re going to be, and a lot of people that are also skeptical of them. I honestly can’t even give you a yes or no answer of how the hell it’s going to pan out. I really don’t know.
Nadav Hollander: 27:07 But if you buy into that thesis, which is a very reasonable thesis that basically bringing all of the efficiencies of crypto tokens into securities trading and clearance is going to totally change the game of how the financial system works, then any robust financial system needs prime brokerage services. It needs somebody that’s basically willing to give you leverage on your assets in some capacity or let you do things like take short positions, etc. That’s basically, when you distill what that business is, it’s basically a relaying business if you think about it. It’s essentially just matching borrowers and lenders in some capacity, though in the traditional financial system that capacity is very custodial. You’re literally getting debt capital from somebody and then lending that out or using that to purchase assets that you’re lending out.
Nadav Hollander: 27:55 But in the relayer model, you could basically build a website that does this entirely non-custodially and has the correct compliance clearances and broker dealer licenses, ATS licenses, etc., that are necessary in order to facilitate all sorts of debt-related transactions in the world of security tokens, essentially. So if somebody wants to, for instance, go short on a CDS that is ensuring some class of sovereign bonds and it’s represented by a token, then they’ll want to borrow that token and somebody is going to need to be able to find the other side of that trade, and that’s where that relayer would come in.
Nadav Hollander: 28:33 That’s one business that we’re pretty excited about that we’d love to see people build out. I’ll give an example from the underwriter side of things, because I think there’s a couple of really interesting niches here. One’s a little bit of a … Sounds silly at first blush, but I actually think this is a really big deal. We refer to it as an appraiser for crypto collectibles. What that means is I’m sure that your listeners are aware of this emerging class of tokens that we refer to as non-fungible crypto collectibles.
Nadav Hollander: 29:06 The most notable example that’s been in the press is CryptoKitties, but this asset class has grown just incredibly rapidly, to the point where … I don’t know if you’re familiar with the website OpenSea, but they’re a marketplace for these different crypto collectibles. This asset class is less than a year old. CryptoKitties was a phenomenon in October, and there’s already 700, 000+ different crypto collectibles listed on this website. It’s wild, right?
Nadav Hollander: 29:36 And you have mainstream brands that are starting to enter the picture. Steph Curry has a CryptoKitty that is themed after him and is co-sponsored by him. There’s all sorts of brands are starting to enter and starting to realize that this is a really cool way of basically marketing themselves with trading cards, functionally. It’s a really interesting emerging asset class. My personal thesis is that this is going to grow tremendously over the coming years, espe-
Nadav Hollander: 30:00 Personal thesis is that this is going to grow tremendously over the coming years, especially as it enters the world of gaming. Many of these assets will be tremendously valuable. On the other hand, by nature of them being like non fungible, they’re not going to be super liquid, right? It’s harder to take a CryptoKitty. It’s not like when you have like an ether, like there’s a spot price for it that you could always find basically a buyer for, right? You could always sell it pretty much instantaneously. While with something CryptoKitty, you basically have to put it up for auction. Essentially, since it is an asset that can become valuable and also is relatively illiquid, it’s actually a great candidate for being used as collateral in alone. It sounds crazy, but I think that there’s a real business here to be made where in order to have things like crypto collectibles be used for leverage for literally for borrowing against them in some capacity, you need to have somebody appraise the value of that asset, right? Somebody needs to come in and be able to determine like, “Well what would the going rate of this CryptoKitty be if it were to be collected on as collateral?” That, in my eyes, is an act of underwriting.
Clay Collins: 31:13 That’s super interesting.
Nadav Hollander: 31:14 Yeah, and so it sounds silly, but I do think that it actually makes a lot of economic sense. The numbers kind of back it up in terms of the volume that’s been going into this niche.
Clay Collins: 31:25 Just thinking about a real world analog, there aren’t really deep markets for fine art for example, or for Van Gogh paintings. If you put them up as collateral, they do have real value and the cost of losing those assets is fairly high. There’s motivation for people to service their debts when that is a collateral. That absolutely makes sense. What about in terms of just super low hanging fruits? I imagine if you built a 0x relayer and you want to offer margin loans, you already understand the model. It’s probably a pretty, I don’t want to say easy, because nothing’s easy in this space, but it’s a logical extension. You understand the model. You understand solidity and how theorem and works. Can you extend that business a bit by offering margin loans with Dharma alongside trades on 0x?
Nadav Hollander: 32:13 I would say that the answer to that is absolutely. To any of the relayers in the 0x ecosystem that are listening right now, we would love to talk to you.
Clay Collins: 32:22 What about factoring? That was something that we talked about prior to this interview is that also seems like low hanging fruit potentially.
Nadav Hollander: 32:30 I think factoring is interesting. The biggest thing with factoring is can you in some way tokenize the expected cash flows that are going to be coming? Can you in some way collateralize those? If it’s just a matter of underwriting, from the underwriter’s perspective, we could imagine underwriters emerging in the Dharma ecosystem for virtually any type of debt, right? The use cases that in my opinion makes the most sense are the ones that are proximate to the crypto industry where people are very expectant to receive some sort of crypto asset or an are comfortable with receiving like Ether die or whatever. With respect to factoring …
Clay Collins: 33:10 Just for a second, could you explain what factoring is for folks who don’t know what that is?
Nadav Hollander: 33:15 Factoring is like invoices. It’s basically like, there is like $100,000 that is en route to me soon that is a paid out in some sort of invoice. It’s for sure coming to me, but I need money right now to do something. Because of, I don’t know, some issues with cross border settlement or something like that, it’s gonna take a while to get to me, but I need cash right now for buying some more inventory or something like that. Factoring is generally loans that are backed by those future cash flows that are essentially somebody gives me a loan right now, so I can buy the inventory that I need. If I don’t pay them back, then they get the cash flows that were coming to me in that invoice. It’s generally considered to be one of the lowest risk classes of debt.
Nadav Hollander: 33:57 In a super crypto-native environment, it’s not super relevant, because inherently crypto is just like super fast, right? It’s basically settles instantly. In a purely crypto environment, it doesn’t map super well. One particular use case that is at least adjacent to that, which I think is pretty interesting, is essentially like, are you familiar with OpenBazaar?
Clay Collins: 34:21 Yes.
Nadav Hollander: 34:22 Yeah, so for those who don’t know, OpenBazaar is this decentralized platform for marketplaces that initially were powered entirely by Bitcoin, but now are powered by, basically larger variety of assets. What’s cool about being a seller in something like OpenBazaar is that you can actually have all of your cash flows that are coming in beyond chain. You actually, in some sense, have like a pretty empirically measurable notion of credit worthiness, because you know what the seller’s public key is. You can see all of the funds that have basically come in there from different public keys. You can see their inventory dwindling over time. I think there’s a really interesting opportunity for essentially like underwriting inventory loans for sellers on places like OpenBazaar. More importantly, I think there’s actually a world in which you could literally like tokenize the expected future cash flows of an OpenBazaar seller and put them up as collateral. You can imagine the address that I have that’s publicly listed, that is supposed to be the one where I receive all of my payments for different inventory items that are purchased on my OpenBazaar store is actually like a tokenized asset that if I don’t pay back a certain loan that I used to buy my inventory, then that address is seized, and now all future profits that come to are basically held as collateral. There’s lots of interesting arrangements you can make.
Clay Collins: 35:45 I just want to riff here where you, because several come to mind. One is if you’re a large employer, and so that’s an example of recurring payments at scale. If someone can commit to locking down a smart contract address through which they’re willing to be paid, maybe in die or some kind of stable coin, and you have to wait 30 days before changing that, then a service they could offer is very low interest, payday loans against a forthcoming paycheck, or if there was a crypto native version of Patreon, if you know what the churn is on that subscription base and you’re willing again to be paid through a smart contract, it’s not perfect.
Clay Collins: 36:25 If you have enough of a track record, it’s very unlikely that your entire supporter base is going to unsubscribe on a given day. If you have 7% monthly churn, maybe you can borrow up to a third of your expected income or a half of your expected income on a given month and see that paid. Another example is maybe a Stripe or a crypto version of Stripe that caters to, or like a Recurly that caters to software as a service businesses. Again, if you have 20,000 customers, your monthly churn is 2%. I think it would be pretty low risk to offer some portion of expected annual income to that business at any given time, again, funneled through the smart contract. It does seem like this is rife with opportunities, but it does require a crypto native infrastructure to really work.
Nadav Hollander: 37:15 Yes, but also you actually brought this example up yourself, which I think is super notable is that it doesn’t necessarily have to be a cash flow that is perfectly tokenizable or like perfectly held as collateral. You mentioned Stripe for instance, which is like a payment processor. They have a trove of data about what the recurring revenue of a whole set of internet businesses looks like. I’m not saying this is likely to happen in the near future, but a Stripe would be an excellent candidate for an underwriter in our protocol insofar as like they have like this perfect image of what’s going on there. Yeah, sure. They could go ahead and like pull a Square and like try to spin up like Stripe Capital and start offering loans, but creating a full lending business is a really onerous activity that requires like a really upfront investment. I would argue that was something like Dharma, you can immediately apply the data science knowledge that they have in house already, probably, just to map out their own personal profitability and use that to underwrite loans.
Clay Collins: 38:15 It’s a great way to monetize your data without putting capital at risk or confusing your investors or your CFO who all of a sudden are going to see new line items on the P and L and wonder what’s going on there, especially if you’re at a VC backed company and you’re already borrowing money in the first place, to enter in a completely different business is absurd. No, I absolutely see it for a company like Patron, essentially any recurring payment business and that includes employers where they have a really good sense of turnover time, I think can provide fantastic data that would allow them to be great underwriters. I liked that a lot.
Nadav Hollander: 38:50 It’s so funny that you mentioned the employer portion of it, because that’s actually something that we are dealing with right now. Literally as of yesterday, we have a certain employee that’s like, they’re looking for an advance on a certain part of their salary, because they’re having a certain issue that they have to deal with. It’s not a big deal. Basically, they’re dealing with this issue right now. We’re literally like trying to arrange this exact thing right now. We could literally use the tech that we’ve been building for the past year and a half to do this.
Clay Collins: 39:19 We’ve now concluded chapter four and use cases for Dharma and programmable tokenized debt. Let’s dive into chapter five where we discussed Dharma as a corporate entity, how it works, where the company’s located, and how they make money. Nadav, can you tell us just a little bit about where you’re based? Are you hiring globally? What kind of legal entity are you? Have you done a token sale or not? Also, can you tell us a little bit about your business operations? Are you a decentralized DOW? How do you operate and function, all that stuff?
Nadav Hollander: 39:48 This is probably going to be the least interesting portion of the Podcast. We are a vanilla Delaware C Corp. We are a for profit company. We do intend to turn a profit one day off of the infrastructure that we built though it is right now, entirely free and open source to use. That will never not be the case. We did not opt to do a token sale, because we felt that though tokens are super promising model, we felt that it wasn’t super lucidly clear how and when they actually capture value. We didn’t think it would be honest to sell basically like tokens to investors that we didn’t even really necessarily have a good explanation of whether they would act as a good proxy for the value we’re building.
Nadav Hollander: 40:29 As for more information about the company, we are based out of San Francisco. Currently, we’re a team of six. We’re rapidly growing. We are very much hiring right now. We’re looking for engineers obviously, but also for marketing people. We’re looking for product people. We’re looking for designers. We’re looking for people who can engage with developers as well. If you fit into any of those categories, we’d love to hear from you. We are primarily hiring locally. We’re looking for people who either are in the San Francisco Bay area or are willing to move out here. We are also open to the possibility of remote hires for particularly exceptional people. If any of the things that I just said in the past two minutes or two hours sounded appealing to you, then we’d love to hear from you.
Clay Collins: 41:15 How are you funded? I know you went through Y Combinator, but that’s certainly not enough to fund six people in a San Francisco office. Have you done a series A?
Nadav Hollander: 41:23 We have not yet done a series A. We raised a seed round approximately almost a year ago at this point now.
Clay Collins: 41:31 If folks want to contact you, either an entrepreneur who is interested in building an app and would like perhaps some guidance or a potential employee who’d like to apply for a position, what’s the best way to get a hold of Dharma or you specifically about those things if you’d rather that be the case?
Nadav Hollander: 41:51 There’s a couple of ways you can find me. First of all, my email is firstname.lastname@example.org. I am super receptive to cold emails though I can’t promise I’ll answer quickly. It’s pretty easy to reach us on our telegram too. If you of go to our website and navigate your way to the telegram channel, you’ll be able to chat with us. Finally, if you’re more explicitly interested in joining the team in some capacity, we have a careers page that you can find our website and apply through there.
Clay Collins: 42:30 We’ve now concluded part two of our deep dive on the Dharma Protocol and tokenized programmable debt. I hope you’ve enjoyed this conversation as much as I have. If you’ve gotten something from it, I hope you share it with others to the degree that you found it useful. Just to close this out, I want to remind you that there’s been a lot of talk in the blockchain space about tokenized stores of value, like cash as well as tokenized securities and commodities, but the topic of tokenized debt that we cover here has received such little coverage despite debt markets being larger than stock markets and money supplies. I just think it needs a lot more attention. I hope this deep dive will contribute to an uptake in dialogue and actual work on tokenized debt markets and work on top of the Dharma Protocol. Most of all, I hope you personally found this content to be helpful. Let’s sign out now. We’re done. See you next week.
Clay Collins: 43:22 That’s it for this week. To sign up for a free crypto investing newsletter, listen to other episodes or get the show notes from this episode, please visit flippening. com. I also invite you to check out the startup that funds this Podcast, Nomics, spelled N-O-M-I-C-S, at nomics.com. Finally, if you got value from the show, the biggest thing you can do to help us out is to leave a five star review with some comments and feedback on iTunes, Stitcher, or wherever you listen to Podcasts. Thanks for listening and see you next week.